An interactive segmentation method for medical images

The interactive segmentation method is a popular technique in medical image processing, since it can obtain more accurate results and minimal manual operations. This paper presents a novel interactive segmentation method, live wire algorithm based on graph searching. Given a seed point, it can obtain and display the boundary of organ, tumor, or vessel tracing the movement of cursor. In addition, some auxiliary functions are introduced to reduce user intervention and obtain more ideal results. Experimental results show that the proposed method is more accurate, efficient, and reproducible than the traditional manual method.